Stars and Planets

< < Module 5 > >

—Birth—

Chris Ormel

Roadmap module 5

Initial Mass Function (IMF)

The initial distribution of stars (after their formation) by their mass

Virial Theorem

A fundamental relation between kinetic/thermal energy and gravitational energy for "relaxed" systems of many particles (stars as well as gas). Frequently employed by astronomers to find the "dynamical mass".

 

Jeans Mass

a critical mass above which the cloud will collapse

 

Dispersion relationship

A relation how a fluid responds to perturbations on a certain scale. A frequently used tool to investigate whether fluids are stable

|

Protoplanetary Disks

The environment where planets are born

Disk instability and core accretion

Planet formation models that describe the formation of terrestrial planets, the cores of giant planets, and gas giants

  • Toomre Q criterion
  • Giant planet formation
  • planetesimal formation
Class 0, 1, 2 and debris disk
  • Coagulation
  • Gravitational focusing
  • Pebble accretion

Virial Theorem and Star Formation

  • Virial theorem
  • Cloud collapse
  • Cloud fragmentation

Virial theorem

— read CO 2.4 —

 

  • The Virial Theorem postulates that for self-gravitating systems:
    • for an N-particle system;
    • for gaseous objects

    where nvir is a positive number, T is the kinetic, Uint the internal energy and W the total gravitational (potential) energy and <..> denotes the time-average. In other words, the virial theorem holds when the system has relaxed into a steady configuration.


    Proof goes along these lines. The quantity

    time-averages to 0. Q (the "virial") is itself the derivative of the moment of inertia. For a gravitating gas, we derive from the hydrostatic balance equation

    To proceed, must be linked to the internal energy per unit mass . For an ideal gas this is just

    where is the number of free degrees and is the heat capacity ratio. For a mono-atomic gas, , and .

Virial theorem

— read CO 2.4 —

 

  • The Virial Theorem postulates that for self-gravitating systems:
    • for an N-particle system;
    • for gaseous objects

    where nvir is a positive number, T is the kinetic, Uint the internal energy and W the total gravitational (potential) energy and <..> denotes the time-average. In other words, the virial theorem holds when the system has relaxed into a steady configuration.

  • Specific examples include:
    •   (fully ionized or mono-atomic gas)
    •   (radiation gas)

Virial theorem

— read CO 2.4 —

 

  • The Virial Theorem postulates that
    • for an N-particle system;
    •   (fully ionized or mono-atomic gas)
    •   (radiation gas)
  • The virial theorem is not the same as energy conservation!

    Conservation of energy dictates that where the total energy is conserved in an isolated system. Hence we obtain that objects in virial equilibrium obey (a different prefactor may apply if the gas is not mono-atomic).
    Note — the expression for the virial theorem can be extended to include magnetic support, rotational support etc.

Virial theorem

— read CO 2.4 —

 

  • The Virial Theorem postulates that
    •   (fully ionized or mono-atomic gas)
    •   (radiation gas)
  • The virial theorem is not the same as energy conservation!
  • Applications:
    • (Proto-)stars heat up when they contract:

      Protostars (no nuclear burning) lose energy. Of the gravitational energy liberated during contraction, half is radiated and the other half goes into internal energy. Stars heat up while loosing energy!
      This corresponds to moving from state A → B in the figure right

    • Stars are thermodynamically stable

      They expand and cool down when their total energy E increases (!). This corresponds to moving from B → A in the figure right. This kind of "virial stability" only applies to the ideal gas law.

    • Obtain the (dynamical) mass by measuring the linewidth.
Dark clouds — by measuring the gas velocity dispersion (linewidth) we obtain a measure of the mass of the cluster.
Globular clusters — by measuring the stellar velocity dispersion (rms-velocity) we obtain a measure of the mass of the cluster.

Virial Theorem — applications

Andromeda galaxy. (c) Peter Forister

Andromeda-Milky Way merger

The Andromeda galaxy will be merging with the Milky Way galaxy in a few billion years. Assume the galaxies are equal in their properties (mass, size, rotation velocity) and that the merger product is again a spiral galaxy.


By the virial theorem, what is the rotational velocity of the merger galaxy?

  1. lower than the present-day
  2. the same
  3. higher than the present-day

Also, what is the radius R of the merger galaxy?

  1. slightly smaller than the present size
  2. the same as the present size
  3. slightly larger than the present
  4. slightly smaller than 2x the present
  5. twice that of the present
  6. larger than 2x the present

Star formation

— read CO 12.2 —

 

The condition for a spherical, uniform cloud to collapse is when its mass exceeds the Jeans mass

When the total energy of the cloud decreases when it is perturbed. If we insert numerical constants for the numbers and write the densities in terms of the number density we obtain

Note some (giant) clouds have suggesting that the should collapse! In reality, these clouds feature additional support mechanisms (like magnetic support). Hence the condition is a necessary but not (always) sufficient condition for collapse.

Star formation

— read CO 12.2 —

 

If the cloud is out of equilibrium, it collapses on a free-fall time

  • During collapse of the cloud, ρ increases and the Jeans mass decreases. The cloud fragments with the fragments collapsing on their own.
  • The isothermal collapse, requires efficient cooling.
  • An increase in temperature, on the other hand, opposes cloud fragmentation, causing MJeans to rise

Star formation

— read CO 12.2 —

 

  • An increase in temperature, on the other hand, opposes cloud fragmentation, causing MJeans to rise

By balancing the heat generated by the collapse with the expression for blackbody radiation applicable when the cloud starts to cool less efficienty ( becomes adiabatic) we can solve for the mass where the fragmentation terminates

where I have omitted numerical factors and ε~0.1 is an efficiency factor. For T=103 K, we obtain .


This is of course an extremely crude estimate. But it tells that the protostar that forms at the end of the collapse state are of stellar mass — not of planet mass or the mass of galaxy. Further accretion onto this protostar will of course occur.

Schematic of how the Jeans mass evolves under collapse of the cloud. Intially, when T is contant, the Jeans mass decreases with density. The cloud therefore fragments. After some point, the heat produced by the collapse (virial theorem!) can no longer escape. In that limit, the collapse becomes adiabatic, T increases and fragmentation is halted.

Star Formation

Initial mass function (IMF)

— read CO p.430 —

The Initial Mass Function or simply IMF is the distribution of stars formed by stellar mass

So ξ(m)Δm gives the number of stars that are formed in mass interval [m,m+Δm]


The IMF follows (steep) power-laws at high masses, but it turns over at low stellar masses (it has to!)


The origin of the IMF is an area of active research in star formation

log10 (m/msun)
mξ(m)
~m-1.8
IMF in the solar neighborhood Rana (1987)

Star formation — Classes

(a)
(b)
(c)
(d)
(e)
(f)
(c) M. Hogerheijde after Shu et al. (1987)
  1. Dark cores form and become unstable when their gravitational potential energy dominates thermal energy:
    (mass exceeds Jeans mass)
  2. The collapse takes place on the free-fall timescale
    this is the embedded phase Class 0

Star formation — Classes

(a)
(b)
(c)
(d)
(e)
(f)
(c) M. Hogerheijde after Shu et al. (1987)
  1. A star has formed. Class 1. An outflow develops, while material is still accreting onto the star. A disks forms.
  2. Accretion and outflow weaken. This is the stage where planets are thought to form. Class 2 or the T-Tauri phase.
  3. After the gas clears, a debris disks may be left

    Debris disk consist of solid material that did not assemble into planetary bodies. In the solar system the asteroid belt and the Kuiper belt are debris disks

  4. A planetary system emerges. The final architecture of the planet system is set when orbital stability is reached.

Dispersion Analysis

  • Linear stability analysis
  • Dispersion relationship for clouds and disks
  • Toomre-Q criterion
  • Giant planet formation and planetesimal formation

Dispersion relation

The fluid equations:

where the E.o.S is chosen as isothermal (constant temperature), such that cs — the isothermal sound speeed — is constant.

To 0th order, for a uniform cloud, the solutions to these equations read simply:

in a dispersion analysis, we consider how wave-like perturbations across a scale λ (or spatial frequency k) superimposed on a "0th-order" background solution fare: do they remain wave-like (positive ω2) or will they grow (negative ω2)?

Dispersion relation

A linearly stability analysis quantifies how the system reacts to small perturbations that are wave-like in nature, e.g.,

with k the wavenumber and ω the frequency (or growth rate when it is imaginary!). Without loss of generality we have assumed that the wave travels in the x-direction. Note than ω can be complex, so the wave can grow (or decay)!


After some algebra, you arrive at the perturbation equations

in a dispersion analysis, we consider how wave-like perturbations across a scale λ (or spatial frequency k) superimposed on a "0th-order" background solution fare: do they remain wave-like (positive ω2) or will they grow (negative ω2)?

Dispersion relation

The solution to the linearly stability analysis is the dispersion relationship:

A dispersion relationship is a relation between the scale of the perturbation (k) and the growth rate (ω).

We obtain:

  • small scales (large k): ω2 is positive: perturbations remain wave-like and do not grow!
  • large scales (small k): ω2 is negative: perturbations grow exponentially and are unstable!
  • a critical scale

    You can verify that amounts to the Jeans mass, barring a numerical coefficient

Dispersion relation & planet formation

For disks the steady solution reads:

the assumption here is that the disk is thin. It does not need to be Keplerian rotating, but it should satisfy Also, we consider that the disk is axisymmetric (quantities do not depend on φ) and consider radial perturbations only (k oriented in the radial direction).

... and the dispersion relationship is

where is the epicycle frequency. For a Keplerian potential . Note that has a minimum corresponding to the critical wavelength.

Dispersion relation & planet formation

... and the dispersion relationship is

We obtain the following:

  • The most critical wavelength is 
  • The disk becomes gravitationally unstable when the Toomre-Q parameter
  • The corresponding mass is .

    Question:
    Is QT <1 the only criterion for gravitational instability to result in collapse?

Dispersion relation & planet formation

... and the dispersion relationship is

We obtain the following:

  • The most critical wavelength is 
  • The disk becomes gravitationally unstable when the Toomre-Q parameter
  • The corresponding mass is .
  • In addition, gravitational collapse requires fast enough cooling

    Quantitatively . Otherwise the formed clumps will just be sheared apart

Dispersion relation & planet formation

... and the dispersion relationship is

We obtain the following:

  • The most critical wavelength is 
  • The disk becomes gravitationally unstable when the Toomre-Q parameter QT<1.
  • In addition, gravitational collapse requires fast enough cooling
  • For dust particles, can be used, amounting to a critical scale of

Hence, particles need to settle. The settling of particles is promoted by the stellar gravity, but opposed by turbulence.

in order to form planetesimals by the Goldreich-Ward mechanisms, solid particles need to settle into a thin layer at the disk midplane.

For the 2D analysis to remain valid, we must have (thin disks). In other words, particles must settle into a thin layer. This is the classical mechanism to form km-sized bodies (planetesimals), also known as the Goldreich-Ward mechanism, see Goldreich & Ward (1973)  

effects of cooling

Smooth Particle Hydrodynamics (SPH) simulations of gravitational instability in disks

  1. standard (fast cooling)
  2. slow cooling

effects of cooling

Smooth Particle Hydrodynamics (SPH) simulations of gravitational instability in disks

  1. standard (fast cooling)
  2. slow cooling

Protoplanetary disks

What are we looking at?

This is hl tau, observed at mm-wavelengths by ALMA. These wavelengths trace the large dust particles in the disk midplane. Intriguing axisymmetric structure —rings— appear

figure credit wikipedia/ESO

you are looking here at mm-wavelength emission
these wavelengths trace the disk midplane

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
Q: Where is the star in this image?
IRAS 04302 — an example of an edge-on disk. Picture taken with the Hubble Space Telescope (visible wavelength).

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
  • the dust results in an IR-excess

    the total integrated emission is a superposition of several black body curves. The IR-excess is how disks were historically detected

spectral energy distribution. The dusty disk is responsible for an IR-excess.

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
  • the dust results in an IR-excess
Short-wavelengths photons tend to be more easily absorbed by dust grains; their absorption cross section is similar to the geometrical cross section. On the other hand, photons of wavelength much larger than the grain size barely interact with the dust grain. This is the Rayleigh scattering limit.

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
  • the dust results in an IR-excess
  • Minimum-mass solar nebula (MMSN)

    By spreading out the heavy element mass of the planets over rings, Weidenschilling (and Hayashi) obtained a crude prescription of the dust and gas density during the time of the formation of the planets. This prescription — known as the Minimum-Mass Solar Nebula — has many flaws, but serves as a useful benchmark. A frequently-used profile is:


Reconstruction of the early disk surface density from the current position of the planets in the solar system. Weidenschilling (1977)

surface density
Q: How did Weidenschilling get
these numbers?
Weidenschilling was brave
enough to fit a power-law!

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
  • the dust results in an IR-excess
  • Minimum-mass solar nebula (MMSN)
  • disks are observed to accrete onto their host star at rates of  

    Mass and angular momentum are being transported.

Herczeg & Hillenbrand (2008)  — emission lines in the ultra violet (UV). These emission lines, observed frequently around Class 1, 2 sources, betray the presence of accretion.

The emission lines can be modeled, which provides the accretion luminosity Lacc. The accretion luminosity and accretion rate are related through:

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
  • the dust results in an IR-excess
  • Minimum-mass solar nebula (MMSN)
  • disks are observed to accrete onto their host star at rates of  

    Mass and angular momentum are being transported.

Manara et al. (2016)  — measurements of the accretion rate

the protoplanetary disks (observations)

read CO p.437—441 & Ch.23.2

 

Properties

  • disks are naturally formed as byproduct of star formation due to angular momentum conservation
  • disks are usually dusty, obscuring the interior disk midplane regions
  • the dust results in an IR-excess
  • Minimum-mass solar nebula (MMSN)
  • disks are observed to accrete onto their host star at rates of  
  • Accretion terminates over timescales of ~Myr

    A natural explanation is that the gas disk is gone after several Myr. It sets a timescale for the formation of gas giant planets

Fedele et al. (2010)  Fraction of stars showing accretion (gas) and those with a near-IR excess (dust)

 

Elements of planet formation

 

 

 

 

(~μm)
(~mm—cm)
(~km)

 

(~μm)
(~mm—cm)
(~km)
(~0.1 ME)

 

(~μm)
(~mm—cm)
(~km)
(~0.1 ME)
(~1 ME)

 

(~μm)
(~mm—cm)
(~km)
(~0.1 ME)
(~1 ME)
(~10 ME)

 

(~μm)
(~mm—cm)
(~km)
(~0.1 ME)
(~1 ME)
(~10 ME)
(>100 ME)

 

Core accretion and Disk instability

There are two main paradigms for the formation of planets:

  1. disk instability: planets form following a gravitational instability in the gas disk
  2. core accretion: planet formation occurs bottom-up by coagulation (collisions between solid particles)

    —initially, solids stick due to surface forces
    —later, bodies are held together by gravity

    (at intermediate sizes sticking is problematic)

end of module 5

—congrats—